Private beta in development
A second opinion for suspicious messages.
A safe space between suspicion and action. PANFILOV.SPACE is building a calm, AI-assisted scam check for messages, calls, links, and payment requests—showing evidence, unknowns, and an independent next step before you act.
Simulated product preview
See the decision layer.
Explore three fixed examples of the planned experience. Nothing is uploaded, analyzed, or stored in this preview.
MessageScreenshotLinkAudio
Planned privacy control: clear the case when you are finished.
Decision support
Review before acting
Evidence
- Urgency and a financial threat
- An unfamiliar link
- Pressure to act immediately
Inference
The sender may be impersonating a bank.
Unknowns
Whether the sender has any connection to the bank.
Illustrative outcomes, not guarantees. This preview is not a live security analysis.
The problem
Scams create urgency. Good decisions need space.
In the United States, consumers reported $3.5 billion in losses to impersonation scams in 2025. Separately, the FBI IC3 recorded more than 22,000 complaints with AI-related information and more than $893 million in adjusted losses. These figures describe reported U.S. cases—not every scam worldwide, and not proof that every impersonation scam used AI.
Built for real decisions
Built for moments that feel urgent.
The first version is focused on the point where pressure is high and independent verification is still possible.
A message asks for money now
A bank, marketplace, employer, government office, or family member appears to need an urgent payment.
A link asks you to sign in
A convincing email or text wants a password, one-time code, card detail, or identity document.
A caller controls the next step
Someone discourages you from hanging up, checking another channel, or speaking with a person you trust. The voice may be cloned, and a familiar one is no longer proof of identity.
Verify independently
Leave the caller's path. Find the real one.
A convincing voice can be a cloned one, and it can still supply a false callback number. PANFILOV.SPACE is designed to turn uncertainty into a small, independent verification plan—without telling people to trust a black-box score.
01
End the call without sharing more information.
02Find the organization through a statement, card, or official site.
03Ask whether the case or request actually exists.
How it works
A safer pause, before the next step.
Share it
Bring the suspicious message, call recording, screenshot, link, or payment request into one calm space. Most people forward a screenshot rather than retype a message, so screenshots are a first-class input rather than an afterthought.
Understand it
See the pressure tactics, contradictions, known facts, and important unknowns—without a black-box verdict.
Verify it
Follow independent checks or ask a trusted person before you respond, share information, or pay.
Act safely
Pause the request, secure an account, contact the real organization, or preserve evidence for recovery.
Three modes
Support that matches the moment.
Check
A second opinion before you click, reply, disclose information, or send money.
Verify
Independent steps to validate identity, links, claims, and payment instructions.
Recover
A calm sequence for preserving evidence and limiting further harm—not a promise of reimbursement.
Family Circle
A trusted person, without pressure.
Planned Family Circle features let someone invite a trusted person into a difficult decision, while keeping the user in control.
Suspicious requestCalm assessmentOptional trusted-person checkSafer action
Principles
Safety by design.
PANFILOV.SPACE is decision support, not a guarantee and not a replacement for banks, emergency services, or law enforcement.
- Reasons, not unexplained confidence scores
- Evidence, inference, and unknowns kept separate
- Minimal retention and user-controlled deletion
- No sale of submitted personal data
- Clear escalation to official channels
- No promise that every case will be classified correctly
Planned technology
Signals become evidence people can use.
The product roadmap combines multimodal AI for screenshots and attachments, speech recognition for call recordings, and deepfake and manipulated-media detection—indicators of synthetic or cloned voice and of AI-generated imagery—alongside explainable risk analysis and privacy-conscious processing.
Impersonation fraud arrives in the language of the person it targets, and the signals that matter are often idiomatic. Multilingual capability is therefore a product requirement rather than a later expansion, and model selection will follow multilingual evaluation instead of English-only benchmarks.
Detection indicators are treated as evidence to weigh, never as a verdict. Provider choices will follow technical evaluation and available startup credits.
Questions before joining
Know what the private beta is — and is not.
Is PANFILOV.SPACE live today?
Not yet. PANFILOV.SPACE is in private-beta development. The product preview on this page uses fixed examples and does not analyze uploads.
What is the product planned to check?
The roadmap covers suspicious messages, screenshots, links, call recordings, and payment requests, with multimodal and multilingual support introduced only after technical and safety evaluation.
Will it check for deepfakes and AI-generated content?
That is central to the roadmap, not a later addition. Planned checks look for indicators of synthetic or cloned voice in call recordings, and of manipulated or AI-generated content in screenshots, images, and attachments. Those indicators are reported as evidence to weigh, with the uncertainty stated. The product is not designed to declare that a recording or image is definitely a deepfake, or to accuse a named person of fraud.
Does it decide that something is definitely a scam?
No. The planned experience separates observed evidence, model inference, and important unknowns, then suggests independent checks. It is decision support, not a guarantee.
How will personal information be handled?
Privacy-conscious processing and clear case deletion are product requirements. For now, the waitlist collects only research preferences. Do not submit suspicious content, recordings, financial details, or personal documents.
Who is PANFILOV.SPACE being designed for?
The initial focus is global English-speaking consumers and families who want a calm second opinion before clicking, replying, sharing information, or paying.
What happens when I join the private beta list?
You may receive product-research invitations and early-access updates. Joining does not guarantee admission or a launch date, and you can contact us if you no longer want updates.
Private beta
Help shape a safer second opinion.
Join the research list for product interviews and early-access updates. We ask about the situations you want help with—not for suspicious content or financial details.